From 9428208d9f72bfeaf514e2cce5ac01385f2a7109 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Luc=C3=A0s?= <86352901+LucasVbr@users.noreply.github.com>
Date: Fri, 4 Oct 2024 09:23:21 +0200
Subject: [PATCH] feat!: Save the converted dataset

/!\ Need to clear the cache folder
---
 package-lock.json               | 32 --------------------------------
 package.json                    |  1 -
 src/services/data/Data.ts       |  5 ++++-
 src/services/data/NudgerData.ts |  6 ++++--
 src/services/dataset/Dataset.ts | 13 ++++++++++++-
 5 files changed, 20 insertions(+), 37 deletions(-)

diff --git a/package-lock.json b/package-lock.json
index 08938c5..ef24dee 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -15,7 +15,6 @@
         "dmn-moddle": "^10.0.0",
         "dotenv": "^16.4.5",
         "express": "^4.21.0",
-        "JSONStream": "^1.3.5",
         "node-stream-zip": "^1.15.0",
         "tar-stream": "^3.1.7",
         "unzipper": "^0.12.3"
@@ -885,31 +884,6 @@
         "graceful-fs": "^4.1.6"
       }
     },
-    "node_modules/jsonparse": {
-      "version": "1.3.1",
-      "resolved": "https://registry.npmjs.org/jsonparse/-/jsonparse-1.3.1.tgz",
-      "integrity": "sha512-POQXvpdL69+CluYsillJ7SUhKvytYjW9vG/GKpnf+xP8UWgYEM/RaMzHHofbALDiKbbP1W8UEYmgGl39WkPZsg==",
-      "engines": [
-        "node >= 0.2.0"
-      ],
-      "license": "MIT"
-    },
-    "node_modules/JSONStream": {
-      "version": "1.3.5",
-      "resolved": "https://registry.npmjs.org/JSONStream/-/JSONStream-1.3.5.tgz",
-      "integrity": "sha512-E+iruNOY8VV9s4JEbe1aNEm6MiszPRr/UfcHMz0TQh1BXSxHK+ASV1R6W4HpjBhSeS+54PIsAMCBmwD06LLsqQ==",
-      "license": "(MIT OR Apache-2.0)",
-      "dependencies": {
-        "jsonparse": "^1.2.0",
-        "through": ">=2.2.7 <3"
-      },
-      "bin": {
-        "JSONStream": "bin.js"
-      },
-      "engines": {
-        "node": "*"
-      }
-    },
     "node_modules/lodash": {
       "version": "4.17.21",
       "resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.21.tgz",
@@ -1386,12 +1360,6 @@
         "b4a": "^1.6.4"
       }
     },
-    "node_modules/through": {
-      "version": "2.3.8",
-      "resolved": "https://registry.npmjs.org/through/-/through-2.3.8.tgz",
-      "integrity": "sha512-w89qg7PI8wAdvX60bMDP+bFoD5Dvhm9oLheFp5O4a2QF0cSBGsBX4qZmadPMvVqlLJBBci+WqGGOAPvcDeNSVg==",
-      "license": "MIT"
-    },
     "node_modules/toidentifier": {
       "version": "1.0.1",
       "resolved": "https://registry.npmjs.org/toidentifier/-/toidentifier-1.0.1.tgz",
diff --git a/package.json b/package.json
index 596d1d8..05f0751 100644
--- a/package.json
+++ b/package.json
@@ -21,7 +21,6 @@
     "dmn-moddle": "^10.0.0",
     "dotenv": "^16.4.5",
     "express": "^4.21.0",
-    "JSONStream": "^1.3.5",
     "node-stream-zip": "^1.15.0",
     "tar-stream": "^3.1.7",
     "unzipper": "^0.12.3"
diff --git a/src/services/data/Data.ts b/src/services/data/Data.ts
index 20d4aee..d12be15 100644
--- a/src/services/data/Data.ts
+++ b/src/services/data/Data.ts
@@ -1,3 +1,6 @@
-interface Data {}
+interface Data {
+  input: string;
+  output: string;
+}
 
 export default Data;
diff --git a/src/services/data/NudgerData.ts b/src/services/data/NudgerData.ts
index c13f0e4..fc7d7c4 100644
--- a/src/services/data/NudgerData.ts
+++ b/src/services/data/NudgerData.ts
@@ -16,10 +16,12 @@ type RawNudgerData = {
 };
 
 class NudgerData implements Data {
-  barcode: string;
+  input: string;
+  output: string;
 
   constructor(rawData: RawNudgerData) {
-    this.barcode = rawData.code;
+    this.input = rawData.code;
+    this.output = rawData.gs1_country;
   }
 }
 
diff --git a/src/services/dataset/Dataset.ts b/src/services/dataset/Dataset.ts
index 30e26be..1bad306 100644
--- a/src/services/dataset/Dataset.ts
+++ b/src/services/dataset/Dataset.ts
@@ -1,4 +1,4 @@
-import { pipeline } from "node:stream";
+import { pipeline, Transform } from "node:stream";
 import { promisify } from "node:util";
 import CacheService from "../CacheService";
 import FileService from "../FileService";
@@ -22,6 +22,7 @@ class Dataset<D> {
 
   /**
    * Create a new dataset instance
+   * @param dConstructor - The constructor of the data class
    * @param id - The unique identifier of the dataset
    * @param url - The URL of the dataset
    * @param sourceFile - The file name of the dataset in the archive
@@ -62,11 +63,21 @@ class Dataset<D> {
 
     const pipelineAsync = promisify(pipeline);
 
+    const self = this;
+
     console.log(`Download: ${this.url}`);
     await pipelineAsync(
       await FileService.getFileStream(this.url),
       archive.extract(this.sourceFile),
       parser.parse(),
+      new Transform({
+        objectMode: true,
+        transform(chunk: object, _, callback) {
+          const data: D = new self.dConstructor(JSON.parse(chunk.toString()));
+          this.push(JSON.stringify(data) + "\n");
+          callback(null, JSON.stringify(data) + "\n");
+        },
+      }),
       FileService.createWriteStream(this.cachePath)
     );
   }